Digital signal processors have a role to play in smart grid applications such as advanced power analytics,PLC and wireless communications,write Thanh Tran and Gustavo Martinez.
The aim of the'smart grid'is to increase the efficiency of national electrical delivery systems.By incorporating monitoring,analysis,control and communications facilities,it is possible to optimise the performance of the electrical grid,allowing electricity to be delivered more efficiently.An example of a PC-controlled smart grid system that utilises power line communication(PLC)and a smart power and energy(SPE)platform to monitor and analyse information(such as signal quality,fault conditions,transmission losses and power consumption.The system must be capable of real-time signal analysis across the grid and transmitting(via wireless communication)the data collected to substations.Digital signal processors(DSPs)are becoming increasingly important for smart grid applications such as advanced power analytics,PLC and wireless communication.DSPs also enable designers to adapt to changing smart grid standards without the need for hardware redesign.The challenge that exists is to develop single devices that incorporate the required processing elements for optimal system performance.One solution is to combine a floating-point DSP with a general purpose processor(GPP).A range of real-time functions(Table 1)can be performed by the DSP,while a high-level operating system,as well as application processing,can be handled by the GPP(e.g.an ARM core).
The architecture of a typical DSP-based smart grid system is shown in Figure 1.The DSP for smart grid systems should have:
A double-precision floating-point unit Low-latency complex signal processing An efficient fast Fourier transform(FFT)engine
Real-time functions:
Precision power analytics Frequency domain analytics Time domain analytics Metrology Rogowski digital integrator Power line communication
Figure 1.DSP-based smart grid architecture showing the main blocks;a 3-phase analogue front-end,a power-analytic DSP,and a PLC DSP
There are three key components that are essential for smart grid deployment:an integrator to ensure precise power analytics,electric metrology to detect poor power quality,and PLC.
Precision power analytics
Smart grid power analytics requires an integrator to detect RMS current(50Hz or 60Hz)from a current sensor device e.g.a Rogowski coil.A DSP-based digital integrator offers benefits over an analogur approach.Low frequency noise rejection,minimum phase error over the entire frequency band,and wide dynamic range are important attributes of a DSP-based integrator that provide the accurate current detection and power analytics needed to determine grid power quality,particularly under severe conditions(e.g.circuit overloading).Rogowski coils provide a cost-effective alternative to current transformers for measuring current.The measured current,however,needs careful processing to recover the true RMS value and requires integration of the Rogowski output waveform(M(dI/dt).
The principal task of the integrator is to recover the fundamental frequency(60Hz)and reject the higher frequency components.Typically the input is a 500Hz signal(around 8x the fundamental frequency)with an output of 20 x log10(60/500)or-18.4dB.The DSP FFT engine processes the integrator output data and displays it as FFT plots.
Detecting poor power qualityElectric metrology is used to detect power quality abnormalities and ensure smart grid stability.This involves measuring RMS current and voltage,reactive,active and apparent power components,and the power signal frequency.Most smart grids require high precision and fast response times,so a floating-point DSP with a 512-sized double-precision floating-point buffer can take just 2.9μs(1300 cycles at 456MHz)to calculate the RMS value.Similarly,DSPs can use used to identify harmonics using discrete fast-Fourier transforms(DFT)or FFT algorithms.This is important as harmonics can result in overheated transformers,false relay tripping,and noticeable light flickering.
